Discussion
This study compared pre-and postoperative frequency-specific hearing outcomes in patients with COM. Of the total cohort, hearing in all frequencies improved significantly postoperatively in AC and ABG test, as expected. The gain is most prominent at low and middle frequencies. This is consistent with previous studies 2. The normal middle ear pressure gain (a result of ossicular coupling) is frequency-dependent 3. As noted by Choi HG et al. that the mean gain decreases 6 dB or less per octave for frequencies above 1000 Hz 1. The low and middle-frequency improvements may represent a normalization of the ossicular coupling effect following reconstructive surgery. In addition to AC elevation, some patients with COM also have BC elevation. Perforation of the tympanic membrane, disruption or fixation of the ossicular chain, change of round window membrane, tympanosclerosis, and fixation of stapes can all cause an elevation in BC 4. Except for 250 and 4000 Hz, the present study identified that BC improved significantly in the whole cohort. The ossicular chain has a large resonance effect at 500-2000 Hz. Weakness or disappearance of resonance caused by the destruction of the ossicular chain can lead to BC change at 500-2000 Hz, and BC can be improved when the continuity and resonance of the ossicular chain are restored through reconstructive surgery 5-6.
At all frequencies except 2000 Hz, the hearing improvement of patients who underwent tympanoplasty was better than that of patients who underwent canal wall-down tympanoplasty in this study. The difference in efficacy between tympanoplasty and canal wall-down tympanoplasty has typically been ascribed to poor Eustachian tube function and greater disease severity in patients requiring canal wall-down tympanoplasty7. Furthermore, Quaranta et al. claimed that an intact posterior canal wall was considered a predictor of the better hearing result in ossiculoplasty 8. Hearing improvement at 2000Hz was significantly better in the canal wall-down tympanoplasty group than in the tympanoplasty group in our investigation. This was consistent with results in research by Şevik Eliçora S et al2. It was perhaps relevant that the intact ossicular chain has the greatest impact on hearing at 2000 Hz 9. The BC changes results were superior in the canal wall-down tympanoplasty group compared with the tympanoplasty group at 250-2000 Hz. The canal wall-down tympanoplasty group had poorer postoperative BC outcomes compared to the tympanoplasty group at 4000 Hz. Excessive ossicular chain movement, vibration or noise from drilling, or other inner ear disruption are all associated with high-frequency sensorineural hearing loss (e.g. strong suction or thermal damage)10. Generally, canal wall-down tympanoplasty has more drilling time in the middle ear near the stapes and oval window, and the BC seems to be more susceptible to its adverse effect at 4000 Hz.
Sound propagation is influenced by the shape, thickness, and anisotropy of the tympanic membrane 11. Most studies show no statistical difference in postoperative hearing improvement with different repair materials. However, they only evaluated the average pure tone threshold. This would tend to obscure fine differences in a specific frequency 12. The AC and ABG statistical results demonstrated that the hearing improvement in the tragus cartilage group was better from 250 to 1000 Hz, but the temporalis fascia group was significantly better from 2000 to 4000 Hz. The rigid nature of tragus cartilage can effective in preventing retraction and resisting negative pressure in the middle ear cavity13. A study by Gan RZ et al. found that the stapes footplate displacement amplitude obtained by tragus cartilage was greater than that obtained by temporalis fascia in the low frequency areas (250-1000 Hz), with better postoperative hearing results14. Others, in contrast to our findings, claim that stiffer graft materials improve high-frequency sound conduction but worsen low-frequency sound conduction 15. As far as we known, the temporal fascia was mainly used for tympanic membrane reconstruction with normal ventilatory function, while tragus cartilage was especially suitable for Eustachian tube dysfunction, adhesion, recurrence of the case 16. A slight deterioration in BC was observed in the temporalis fascia group at 4000 Hz, but the sensorineural hearing loss was unrelated of the type of graft used during reconstruction 17.
Surprisingly, the TORP group had the best hearing improvement, followed by the PORP group and the None group at all frequencies except 1000 Hz. None group was closer to normal hearing, to begin with, and had less room for improvement. Most studies assumed that PORP was associated with superior postoperative hearing outcomes when compared with TORP1. They suggest that the stapes suprastructure plays an important role in hearing 18, and the stability between TORP and stapes floor is inadequate and prone to displacement, resulting in the interruption of TORP and stapes footplate connection. But in their studies, postoperative audiometry tests were generally conducted about three months after surgery. It is our opinion that the residual stapes in subsequent recovery may continue to be disrupted and adhesion with the surrounding tissue. Furthermore, ossicular coupling of TORP may be superior to that of PORP 19. Cadaveric temporal bone studies support this viewpoint and considered that TORP placement to a stapes footplate offers acoustic advantages over other ossiculoplasty techniques. The reduced middle ear pressure can be restored after ossichoplasty. Therefore, the results suggest that AC changes occurred principally at lower and mid frequencies, with no substantial changes at higher frequencies. Tonndorf et al. believes that the influence of ossicular chain mechanics on BC is expressed at most as the improvement for the frequency of 2000 Hz, which is caused by the reduction or elimination of resonance within the ossicular chain20. Similarly, we discovered that three groups all showed significantly improved at 1000-2000 Hz.